| Age | Commit message (Collapse) | Author |
|---|
|
The ETM perf infrastructure closes out a handle during event_stop
or on an error in starting the event. In either case, it is possible
for a "sink" to update/close the handle, under certain circumstances.
(e.g no space in ring buffer.). So, ensure that we handle this
gracefully in the PMU driver by verifying the handle is still valid.
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Leo Yan <leo.yan@linaro.org>
Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20210914102641.1852544-4-suzuki.poulose@arm.com
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
The Trace Filtering support (FEAT_TRF) ensures that the ETM
can be prohibited from generating any trace for a given EL.
This is much stricter knob, than the TRCVICTLR exception level
masks, which doesn't prevent the ETM from generating Context
packets for an "excluded" EL. At the moment, we do a onetime
enable trace at user and kernel and leave it untouched for the
kernel life time. This implies that the ETM could potentially
generate trace packets containing the kernel addresses, and
thus leaking the kernel virtual address in the trace.
This patch makes the switch dynamic, by honoring the filters
set by the user and enforcing them in the TRFCR controls.
We also rename the cpu_enable_tracing() appropriately to
cpu_detect_trace_filtering() and the drvdata member
trfc => trfcr to indicate the "value" of the TRFCR_EL1.
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Al Grant <al.grant@arm.com>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>
Link: https://lore.kernel.org/r/20210914102641.1852544-3-suzuki.poulose@arm.com
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
When the CPU enters a low power mode, the TRFCR_EL1 contents could be
reset. Thus we need to save/restore the TRFCR_EL1 along with the ETM4x
registers to allow the tracing.
The TRFCR related helpers are in a new header file, as we need to use
them for TRBE in the later patches.
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Leo Yan <leo.yan@linaro.org>
Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20210914102641.1852544-2-suzuki.poulose@arm.com
[Fixed cosmetic details]
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
combos
When a traced process runs on a CPU that can't reach the selected sink,
the event will be stopped with PERF_HES_STOPPED. This means that even if
the process migrates to a valid CPU, tracing will not resume.
This can be reproduced (on N1SDP) by using taskset to start the process
on CPU 0, and then switching it to CPU 2 (ETF 1 is only reachable from
CPU 2):
taskset --cpu-list 0 ./perf record -e cs_etm/@tmc_etf1/ --per-thread -- taskset --cpu-list 2 ls
This produces a single 0 length AUX record, and then no more trace:
0x3c8 [0x30]: PERF_RECORD_AUX offset: 0 size: 0 flags: 0x1 [T]
After the fix, the same command produces normal AUX records. The perf
self test "89: Check Arm CoreSight trace data recording and synthesized
samples" no longer fails intermittently. This was because the taskset in
the test is after the fork, so there is a period where the task is
scheduled on a random CPU rather than forced to a valid one.
Specifically selecting an invalid CPU will still result in a failure to
open the event because it will never produce trace:
./perf record -C 2 -e cs_etm/@tmc_etf0/
failed to mmap with 12 (Cannot allocate memory)
The only scenario that has changed is if the CPU mask has a valid CPU
sink combo in it.
Testing
=======
* Coresight self test passes consistently:
./perf test Coresight
* CPU wide mode still produces trace:
./perf record -e cs_etm// -a
* Invalid -C options still fail to open:
./perf record -C 2,3 -e cs_etm/@tmc_etf0/
failed to mmap with 12 (Cannot allocate memory)
* Migrating a task to a valid sink/CPU now produces trace:
taskset --cpu-list 0 ./perf record -e cs_etm/@tmc_etf1/ --per-thread -- taskset --cpu-list 2 ls
* If the task remains on an invalid CPU, no trace is emitted:
taskset --cpu-list 0 ./perf record -e cs_etm/@tmc_etf1/ --per-thread -- ls
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: James Clark <james.clark@arm.com>
Link: https://lore.kernel.org/r/20210922125144.133872-2-james.clark@arm.com
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
ASUS ProArt X570-CREATOR WIFI board has got an nct6775 chip, but by default
there's no use of it because of resource conflict with WMI method.
This commit adds ProArt X570-CREATOR WIFI to the list of boards that can be
monitored using ASUS WMI.
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=204807
Signed-off-by: Denis Pauk <pauk.denis@gmail.com>
Tested-by: Joel Wirāmu Pauling <jwp@redhat.com>
Link: https://lore.kernel.org/r/20211025194748.6784-1-pauk.denis@gmail.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
The AUX bounce buffer is allocated with API dma_alloc_coherent(), in the
low level's architecture code, e.g. for Arm64, it maps the memory with
the attribution "Normal non-cacheable"; this can be concluded from the
definition for pgprot_dmacoherent() in arch/arm64/include/asm/pgtable.h.
Later when access the AUX bounce buffer, since the memory mapping is
non-cacheable, it's low efficiency due to every load instruction must
reach out DRAM.
This patch changes to allocate pages with dma_alloc_noncoherent(), the
driver can access the memory via cacheable mapping; therefore, load
instructions can fetch data from cache lines rather than always read
data from DRAM, the driver can boost memory performance. After using
the cacheable mapping, the driver uses dma_sync_single_for_cpu() to
invalidate cacheline prior to read bounce buffer so can avoid read stale
trace data.
By measurement the duration for function tmc_update_etr_buffer() with
ftrace function_graph tracer, it shows the performance significant
improvement for copying 4MiB data from bounce buffer:
# echo tmc_etr_get_data_flat_buf > set_graph_notrace // avoid noise
# echo tmc_update_etr_buffer > set_graph_function
# echo function_graph > current_tracer
before:
# CPU DURATION FUNCTION CALLS
# | | | | | | |
2) | tmc_update_etr_buffer() {
...
2) # 8148.320 us | }
after:
# CPU DURATION FUNCTION CALLS
# | | | | | | |
2) | tmc_update_etr_buffer() {
...
2) # 2525.420 us | }
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20210905032144.966766-1-leo.yan@linaro.org
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
Commit 2f01c200d440 ("perf cs-etm: Remove callback cs_etm_find_snapshot()")
has removed the function cs_etm_find_snapshot() from the perf tool in the
user space, now CoreSight trace directly uses the perf common function
__auxtrace_mmap__read() to calcualte the head and size for AUX trace data
in snapshot mode.
This patch updates the comments in drivers to make them generic and not
stick to any specific function from perf tool.
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Link: https://lore.kernel.org/r/20210912125748.2816606-3-leo.yan@linaro.org
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
When enable the Arm CoreSight PMU event, the context for AUX ring buffer
is prepared in the structure perf_output_handle, and its field "head"
points the head of the AUX ring buffer and it is updated after filling
AUX trace data into buffer.
Current code uses an extra field etr_perf_buffer::head to maintain the
header for the AUX ring buffer which is not necessary; alternatively,
it's better to directly use perf_output_handle::head.
This patch removes the field etr_perf_buffer::head and directly uses
perf_output_handle::head for the head of AUX ring buffer.
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20210912125748.2816606-2-leo.yan@linaro.org
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
Since the function CS_LOCK() has contained memory barrier mb(), it
ensures the visibility of the AUX trace data before updating the
aux_head, thus it's needless to add any explicit barrier anymore.
Add comment to make clear for the barrier usage for ETF.
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Link: https://lore.kernel.org/r/20210809111407.596077-4-leo.yan@linaro.org
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
Since a memory barrier is required between AUX trace data store and
aux_head store, and the AUX trace data is filled with memcpy(), it's
sufficient to use smp_wmb() so can ensure the trace data is visible
prior to updating aux_head.
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Reviewed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20210809111407.596077-3-leo.yan@linaro.org
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
The current driver sets the write burst size initiated by TMC-ETR on
AXI bus to a fixed value of 16. Make this configurable by reading the
value specified in fwnode. If not specified, then default to 16.
Introduced a "max_burst_size" variable in tmc_drvdata structure to
facilitate this change.
Signed-off-by: Tanmay Jagdale <tanmay@marvell.com>
Reviewed-by: Mike Leach <mike.leach@linaro.org>
Link: https://lore.kernel.org/r/20210901131049.1365367-3-tanmay@marvell.com
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
Debugfs is nice and so are module parameters, but
* debugfs doesn't take effect early (e.g., if drivers are locking up
before user space gets anywhere)
* module parameters either add a lot to the kernel command line, or
else take effect late as well (if you build =m and configure in
/etc/modprobe.d/)
So in the same spirit as these
CONFIG_PANIC_ON_OOPS (also available via cmdline or modparam)
CONFIG_INTEL_IOMMU_DEFAULT_ON (also available via cmdline)
add a new Kconfig option.
Module parameters and debugfs can still override.
Signed-off-by: Brian Norris <briannorris@chromium.org>
Reviewed-by: Leo Yan <leo.yan@linaro.org>
[Fixed missing double quote in Kconfig title]
Link: https://lore.kernel.org/r/20210903182839.1.I20856983f2841b78936134dcf9cdf6ecafe632b9@changeid
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
The input parameter of the function pm_runtime_put should be the
same in the function cti_enable_hw and cti_disable_hw. The correct
parameter to use here should be dev->parent.
Signed-off-by: Tao Zhang <quic_taozha@quicinc.com>
Reviewed-by: Leo Yan <leo.yan@linaro.org>
Fixes: 835d722ba10a ("coresight: cti: Initial CoreSight CTI Driver")
Cc: stable <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/1629365377-5937-1-git-send-email-quic_taozha@quicinc.com
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
|
|
This change allows LTD and RTD inputs to be configured via
device tree bindings. If the DT bindings are not present or
invalid, the input configuration is not modified and left at
HW defaults.
Signed-off-by: Oskar Senft <osk@google.com>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Link: https://lore.kernel.org/r/20211020164213.174597-2-osk@google.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
|
|
Pull rdma fixes from Jason Gunthorpe:
"Nothing very exciting here, it has been a quiet cycle overall. Usual
collection of small bug fixes:
- irdma issues with CQ entries, VLAN completions and a mutex deadlock
- Incorrect DCT packets in mlx5
- Userspace triggered overflows in qib
- Locking error in hfi
- Typo in errno value in qib/hfi1
- Double free in qedr
- Leak of random kernel memory to userspace with a netlink callback"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
RDMA/sa_query: Use strscpy_pad instead of memcpy to copy a string
RDMA/irdma: Do not hold qos mutex twice on QP resume
RDMA/irdma: Set VLAN in UD work completion correctly
RDMA/mlx5: Initialize the ODP xarray when creating an ODP MR
rdma/qedr: Fix crash due to redundant release of device's qp memory
RDMA/rdmavt: Fix error code in rvt_create_qp()
IB/hfi1: Fix abba locking issue with sc_disable()
IB/qib: Protect from buffer overflow in struct qib_user_sdma_pkt fields
RDMA/mlx5: Set user priority for DCT
RDMA/irdma: Process extended CQ entries correctly
|
|
mt76 patches for 5.16
* fix a compile error with !CONFIG_PM
* cleanups
* MT7915 DBDC fixes
* endian warning fixes
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/phy/linux-phy into char-misc-next
Vinod writes:
phy-for-5.16
- New support:
- Kirin 970 PCIe PHY driver
- Qualcomm QCM2290 USB2 and USB3 support
- Updates:
- Qualcomm synopsis phy driver updates
- sc8180x PCIe update
- cadence-torrent driver updates for output reference clock
- stm32 phy tuning support
* tag 'phy-for-5.16' of git://git.kernel.org/pub/scm/linux/kernel/git/phy/linux-phy: (28 commits)
phy: Sparx5 Eth SerDes: Fix return value check in sparx5_serdes_probe()
phy: qcom-snps: Correct the FSEL_MASK
phy: hisilicon: Add of_node_put() in phy-hisi-inno-usb2
phy: qcom-qmp: another fix for the sc8180x PCIe definition
phy: cadence-torrent: Add support to output received reference clock
phy: cadence-torrent: Model reference clock driver as a clock to enable derived refclk
dt-bindings: phy: cadence-torrent: Add clock IDs for derived and received refclk
phy: cadence-torrent: Migrate to clk_hw based registration and OF APIs
phy: ti: gmii-sel: check of_get_address() for failure
dt-bindings: phy: qcom,qmp: IPQ6018 and IPQ8074 PCIe PHY require no supply
phy: stm32: add phy tuning support
dt-bindings: phy: phy-stm32-usbphyc: add optional phy tuning properties
phy: stm32: restore utmi switch on resume
dt-bindings: phy: rockchip: remove usb-phy fallback string for rk3066a/rk3188
phy: qcom-qusb2: Fix a memory leak on probe
phy: qcom-qmp: Add QCM2290 USB3 PHY support
dt-bindings: phy: qcom,qmp: Add QCM2290 USB3 PHY
phy: qcom-qusb2: Add missing vdd supply
dt-bindings: phy: qcom,qusb2: Add missing vdd-supply
phy: rockchip-inno-usb2: Make use of the helper function devm_add_action_or_reset()
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/extcon into char-misc-next
Chanwoo writes:
Update extcon next for v5.16
Detailed description for this pull request:
1. Add support for TUSB320L and update tusb320 extcon driver
- The existing extcon-usbc-tusb320 driver is updated
for supporting the mode setting and reset operation.
Also, this driver supports the simliar TUSB320L device
at the same extcon-usbc-tusb320 extcon provider driver.
2. Use p-unit semaphone lock for register access for extcon-axp288
driver
3. Update the minor clean-up for extcon-max3355 and extcon-usb-gpio
driver.
* tag 'extcon-next-for-5.16' of git://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/extcon:
dt-bindings: extcon: usbc-tusb320: Add TUSB320L compatible string
extcon: usbc-tusb320: Add support for TUSB320L
extcon: usbc-tusb320: Add support for mode setting and reset
extcon: extcon-axp288: Use P-Unit semaphore lock for register accesses
extcon: max3355: Drop unused include
extcon: usb-gpio: Use the right includes
|
|
Add support for new system type, which is a water-cooling flavor
of the VMOD001 system class, equipped with 48xSFP28 and 8xQSFP28
100G Ethernet ports.
System is recognized by "DMI_BOARD_NAME" and " DMI_PRODUCT_SKU"
matches, when these fields are set respectively to "VMOD001" and
"HI138".
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Reviewed-by: Oleksandr Shamray <oleksandrs@nvidia.com>
Link: https://lore.kernel.org/r/20211023094022.4193813-4-vadimp@nvidia.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
systems
Extend systems of class VMOD0010 equipped with CoffeeLake COMEx module
with BIOS related attributes to represent various BIOS statuses.
These attributes "bios_active_image", "bios_auth_fail",
"bios_upgrade_fail", "bios_safe_mode" has been already added to modular
system. This all of them are already documented.
- "bios_active_image" - location of current active BIOS image (0: Top,
1: Bottom. The reported value should correspond to value expected by
OS in case of BIOS safe mode is 0. This bit is related to Intel
top-swap feature of DualBios on the same flash.
- "bios_auth_fail": BIOS upgrade is failed because provided BIOS image
is not signed correctly.
- "bios_upgrade_fail" BIOS upgrade is failed by some reason not related
to authentication. For example, due to physical SPI flash problem.
- "bios_safe_mod": - 0 : if BIOS is booted from a supposed active image;
1 : BIOS safe mechanism was enforced by hardware (CPLD).
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Reviewed-by: Oleksandr Shamray <oleksandrs@nvidia.com>
Link: https://lore.kernel.org/r/20211023094022.4193813-3-vadimp@nvidia.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
MQM97xx systems
Add support for new system types "MQM97xx", which is based on Mellanox
Quantum-2 ASIC. It provides up to 64x400GB/s (IB) full bidirectional
bandwidth per port using PAM-4 modulation. The system support 32 OSFP
cages that can provide 64x400GB/s per port (two ports/cage). The system
fits standard 1U racks.
System is equipped with seven fan drawers and with per fan drawer LED
on backport panel and uses two-bytes for exposing CPLD Part Number
versions.
System is recognized by "DMI_BOARD_NAME" match, when this field is set
to "VMOD0010".
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Reviewed-by: Oleksandr Shamray <oleksandrs@nvidia.com>
Link: https://lore.kernel.org/r/20211023094022.4193813-2-vadimp@nvidia.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
This reverts commits 55845f46df03 and c50031f03dfe, since this still
causes a build failure when QCOM_SCM is a loadable module, or when
CONFIG_SMP is disabled:
ERROR: modpost: "cpu_logical_map" [drivers/firmware/qcom-scm.ko] undefined!
This be done better for 5.17, but it's too late now to rework
properly.
Fixes: c50031f03dfe ("firmware: qcom: scm: Don't break compile test on non-ARM platforms")
Fixes: 55845f46df03 ("firmware: qcom: scm: Add support for MC boot address API")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
An upcoming change to platform profiles will export `platform_profile_get`
as a symbol that can be used by other drivers. Avoid the collision.
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20211026190835.10697-3-mario.limonciello@amd.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
An upcoming change to platform profiles will export `platform_profile_get`
as a symbol that can be used by other drivers. Avoid the collision.
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20211026190835.10697-2-mario.limonciello@amd.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
This is already checked when calling rtc_read_alarm.
Suggested-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20211026171443.289-3-mario.limonciello@amd.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
For the majority of users this information will not be informative
as they've chosen to program the RTC before going to sleep.
Suggested-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20211026171443.289-2-mario.limonciello@amd.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
Just hardcode the RTC to "rtc0" which is the default for
CONFIG_RTC_SYSTOHC_DEVICE and used by all standard x86 distros.
Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
Suggested-by: Hans de Goede <hdegoede@redhat.com>
Fixes: 59348401ebed ("platform/x86: amd-pmc: Add special handling for timer based S0i3 wakeup")
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20211026171443.289-1-mario.limonciello@amd.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
Now that we have a better method to select SFP interface modes, we
no longer need to use phylink_helper_basex_speed() in a driver's
validation function, and we can also get rid of our hack to indicate
both 1000base-X and 2500base-X if the comphy is present to make that
work. Remove this hack and use of phylink_helper_basex_speed().
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
As phylink checks the interface mode against the supported_interfaces
bitmap, we no longer need to validate the interface mode in the
validation function. Remove this to simplify it.
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Populate the phy_interface_t bitmap for the Marvell mvneta driver with
interfaces modes supported by the MAC.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This patch adjusts the string spaces of some parameters of tx bd info in
debugfs according to their maximum needs.
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The specified buffer length for three debugfs files fd_tcam, uc and tqp
is not enough for their maximum needs, so this patch fixes them.
Fixes: b5a0b70d77b9 ("net: hns3: refactor dump fd tcam of debugfs")
Fixes: 1556ea9120ff ("net: hns3: refactor dump mac list of debugfs")
Fixes: d96b0e59468d ("net: hns3: refactor dump reg of debugfs")
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
in debugfs
As the width of packets number registers is 32 bits, they needs at most
10 characters for decimal data printing, but now the string spaces is not
enough, so this patch fixes it.
Fixes: e44c495d95e ("net: hns3: refactor queue info of debugfs")
Signed-off-by: Jie Wang <wangjie125@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The member data in struct hclge_desc is type of __le32, it needs endian
conversion before using it, and some functions of debugfs didn't do that,
so this patch fixes it.
Fixes: c0ebebb9ccc1 ("net: hns3: Add "dcb register" status information query function")
Signed-off-by: Jie Wang <wangjie125@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Currently, if there is a reset event triggered by RAS during device in
initialization process, driver may run reset process concurrently with
initialization process. In this case, it may cause problem. For example,
the RSS indirection table may has not been alloc memory in initialization
process yet, but it is used in reset process, it will cause a call trace
like this:
[61228.744836] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000
...
[61228.897677] Workqueue: hclgevf hclgevf_service_task [hclgevf]
[61228.911390] pstate: 40400009 (nZcv daif +PAN -UAO -TCO BTYPE=--)
[61228.918670] pc : hclgevf_set_rss_indir_table+0xb4/0x190 [hclgevf]
[61228.927812] lr : hclgevf_set_rss_indir_table+0x90/0x190 [hclgevf]
[61228.937248] sp : ffff8000162ebb50
[61228.941087] x29: ffff8000162ebb50 x28: ffffb77add72dbc0 x27: ffff0820c7dc8080
[61228.949516] x26: 0000000000000000 x25: ffff0820ad4fc880 x24: ffff0820c7dc8080
[61228.958220] x23: ffff0820c7dc8090 x22: 00000000ffffffff x21: 0000000000000040
[61228.966360] x20: ffffb77add72b9c0 x19: 0000000000000000 x18: 0000000000000030
[61228.974646] x17: 0000000000000000 x16: ffffb77ae713feb0 x15: ffff0820ad4fcce8
[61228.982808] x14: ffffffffffffffff x13: ffff8000962eb7f7 x12: 00003834ec70c960
[61228.991990] x11: 00e0fafa8c206982 x10: 9670facc78a8f9a8 x9 : ffffb77add717530
[61229.001123] x8 : ffff0820ad4fd6b8 x7 : 0000000000000000 x6 : 0000000000000011
[61229.010249] x5 : 00000000000cb1b0 x4 : 0000000000002adb x3 : 0000000000000049
[61229.018662] x2 : ffff8000162ebbb8 x1 : 0000000000000000 x0 : 0000000000000480
[61229.027002] Call trace:
[61229.030177] hclgevf_set_rss_indir_table+0xb4/0x190 [hclgevf]
[61229.039009] hclgevf_rss_init_hw+0x128/0x1b4 [hclgevf]
[61229.046809] hclgevf_reset_rebuild+0x17c/0x69c [hclgevf]
[61229.053862] hclgevf_reset_service_task+0x4cc/0xa80 [hclgevf]
[61229.061306] hclgevf_service_task+0x6c/0x630 [hclgevf]
[61229.068491] process_one_work+0x1dc/0x48c
[61229.074121] worker_thread+0x15c/0x464
[61229.078562] kthread+0x168/0x16c
[61229.082873] ret_from_fork+0x10/0x18
[61229.088221] Code: 7900e7f6 f904a683 d503201f 9101a3e2 (38616b43)
[61229.095357] ---[ end trace 153661a538f6768c ]---
To fix this problem, don't schedule reset task before initialization
process is done.
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Currently, the workqueue of hclge/hclgevf is executed on
the CPU that initiates scheduling requests by default. In
stress scenarios, the CPU may be busy and workqueue scheduling
is completed after a long period of time. To avoid this
situation and implement proper scheduling, use the WQ_UNBOUND
mode instead. In this way, the workqueue can be performed on
a relatively idle CPU.
Signed-off-by: Yufeng Mo <moyufeng@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
If a TP port is configured by follow steps:
1.ethtool -s ethx autoneg off speed 100 duplex full
2.ethtool -A ethx rx on tx on
3.ethtool -s ethx autoneg on(rx&tx negotiated pause results are off)
4.ethtool -s ethx autoneg off speed 100 duplex full
In step 3, driver will set rx&tx pause parameters of hardware to off as
pause parameters negotiated with link partner are off.
After step 4, the "ethtool -a ethx" command shows both rx and tx pause
parameters are on. However, pause parameters of hardware are still off
and port has no flow control function actually.
To fix this problem, if autoneg is disabled, driver uses its saved
parameters to restore pause of hardware. If the speed is not changed in
this case, there is no link state changed for phy, it will cause the pause
parameter is not taken effect, so we need to force phy to go down and up.
Fixes: aacbe27e82f0 ("net: hns3: modify how pause options is displayed")
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
I got memory leak as follows when doing fault injection test:
unreferenced object 0xffff888258228440 (size 64):
comm "kworker/7:2", pid 2005, jiffies 4294989509 (age 824.540s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
backtrace:
[<ffffffff8167939c>] slab_post_alloc_hook+0x9c/0x490
[<ffffffff8167f627>] kmem_cache_alloc_trace+0x1f7/0x470
[<ffffffffa02ac0e4>] keyspan_port_probe+0xa4/0x5d0 [keyspan]
[<ffffffffa0294c07>] usb_serial_device_probe+0x97/0x1d0 [usbserial]
[<ffffffff82b50ca7>] really_probe+0x167/0x460
[<ffffffff82b51099>] __driver_probe_device+0xf9/0x180
[<ffffffff82b51173>] driver_probe_device+0x53/0x130
[<ffffffff82b516f5>] __device_attach_driver+0x105/0x130
[<ffffffff82b4cfe9>] bus_for_each_drv+0x129/0x190
[<ffffffff82b50a69>] __device_attach+0x1c9/0x270
[<ffffffff82b518d0>] device_initial_probe+0x20/0x30
[<ffffffff82b4f062>] bus_probe_device+0x142/0x160
[<ffffffff82b4a4e9>] device_add+0x829/0x1300
[<ffffffffa0295fda>] usb_serial_probe.cold+0xc9b/0x14ac [usbserial]
[<ffffffffa02266aa>] usb_probe_interface+0x1aa/0x3c0 [usbcore]
[<ffffffff82b50ca7>] really_probe+0x167/0x460
If keyspan_port_probe() fails to allocate memory for an out_buffer[i] or
in_buffer[i], the previously allocated memory for out_buffer or
in_buffer needs to be freed on the error handling path, otherwise a
memory leak will result.
Fixes: bad41a5bf177 ("USB: keyspan: fix port DMA-buffer allocations")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wang Hai <wanghai38@huawei.com>
Link: https://lore.kernel.org/r/20211015085543.1203011-1-wanghai38@huawei.com
Cc: stable@vger.kernel.org # 3.12
Signed-off-by: Johan Hovold <johan@kernel.org>
|
|
The new wrapper functions for usb_control_msg() can accept data from
stack and treat short reads as errors. Hence use the wrappers functions.
Please note that because of this change, cp210x_read_reg_block() will no
longer log the length of short reads.
Signed-off-by: Himadri Pandya <himadrispandya@gmail.com>
Link: https://lore.kernel.org/r/20211001065720.21330-3-himadrispandya@gmail.com
[ johan: style changes ]
Signed-off-by: Johan Hovold <johan@kernel.org>
|
|
usb_control_msg_recv() is a new wrapper function for usb_control_msg()
that has error checks for short reads. This function also accepts data
buffer on stack. Hence use this function to simplify error handling for
short reads. Short reads will now get reported as -EREMOTEIO with no
indication of how short the transfer was.
Signed-off-by: Himadri Pandya <himadrispandya@gmail.com>
Link: https://lore.kernel.org/r/20211001065720.21330-2-himadrispandya@gmail.com
[ johan: fix quirk-detection breakage, style changes ]
Signed-off-by: Johan Hovold <johan@kernel.org>
|
|
There was nothing to protect multiple SPI controllers on the same FSI2SPI
device from being accessed through the FSI2SPI device at the same time.
For example, multiple writes to the command and data registers might occur
for different SPI controllers, resulting in complete chaos in the SPI
engine. To prevent this, add a FSI2SPI device level mutex and lock it in
the SPI register read and write functions.
Fixes: bbb6b2f9865b ("spi: Add FSI-attached SPI controller driver")
Signed-off-by: Eddie James <eajames@linux.ibm.com>
Link: https://lore.kernel.org/r/20211026193327.52420-1-eajames@linux.ibm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The wait_for_completion_timeout function returns 0 if timed out or a
positive value if completed. Hence, "less than zero" comparison always
misses timeouts and doesn't kill the URB as it should, leading to
re-sending it while it is active.
Fixes: 42337b9d4d95 ("HID: add driver for U2F Zero built-in LED and RNG")
Signed-off-by: Andrej Shadura <andrew.shadura@collabora.co.uk>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
|
|
The previous commit fixed handling of incomplete packets but broke error
handling: offsetof returns an unsigned value (size_t), but when compared
against the signed return value, the return value is interpreted as if
it were unsigned, so negative return values are never less than the
offset.
To make the code easier to read, calculate the minimal packet length
once and separately, and assign it to a signed int variable to eliminate
unsigned math and the need for type casts. It then becomes immediately
obvious how the actual data length is calculated and why the return
value cannot be less than the minimal length.
Fixes: 22d65765f211 ("HID: u2fzero: ignore incomplete packets without data")
Fixes: 42337b9d4d95 ("HID: add driver for U2F Zero built-in LED and RNG")
Signed-off-by: Andrej Shadura <andrew.shadura@collabora.co.uk>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
|
|
NitroKey produced a clone of U2F Zero with a different firmware,
which moved extra commands into the vendor range.
Disambiguate hardware revisions and select the correct configuration in
u2fzero_probe.
Link: https://github.com/Nitrokey/nitrokey-fido-u2f-firmware/commit/a93c16b41f
Signed-off-by: Andrej Shadura <andrew.shadura@collabora.co.uk>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
|
|
The helper function devm_add_action_or_reset() will internally
call devm_add_action(), and if devm_add_action() fails then it will
execute the action mentioned and return the error code. So
use devm_add_action_or_reset() instead of devm_add_action()
to simplify the error handling, reduce the code.
Signed-off-by: Cai Huoqing <caihuoqing@baidu.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
|
|
The size of the critical section in this function appears to be larger
than necessary. The `wacom_udev_list_lock` exists to ensure that one
interface cannot begin checking if a shared object exists while a second
interface is doing the same (otherwise both could determine that no
object exists yet and create their own independent objects rather than
sharing just one). It should be safe for the critical section to end
once a fresly-allocated shared object would be found by other threads
(i.e., once it has been added to `wacom_udev_list`, which is looped
over by `wacom_get_hdev_data`).
This commit is a necessary pre-requisite for a later change to swap the
use of `devm_add_action` with `devm_add_action_or_reset`, which would
otherwise deadlock in its error case.
Signed-off-by: Jason Gerecke <jason.gerecke@wacom.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
|
|
DART has an additional global register to control which streams are
isolated. This register is a bit redundant since DART_TCR can already
be used to control isolation and is usually initialized to DART_STREAM_ALL
by the time we get control. Some DARTs (namely the one used for the audio
controller) however have some streams disabled initially. Make sure those
work by initializing DART_STREAMS_ENABLE during reset.
Reported-by: Martin Povišer <povik@protonmail.com>
Signed-off-by: Sven Peter <sven@svenpeter.dev>
Reviewed-by: Hector Martin <marcan@marcan.st>
Link: https://lore.kernel.org/r/20211019162253.45919-1-sven@svenpeter.dev
Signed-off-by: Joerg Roedel <jroedel@suse.de>
|
|
This patch adds a check for if the rumble queue ringbuffer is empty
prior to queuing the rumble workqueue. If the current rumble setting is
using a non-zero amplitude though, it will queue the worker anyway. This
is because the controller will automatically disable the rumble effect
if it isn't "refreshed".
This change improves bluetooth communication reliability with the
controller, since it reduces the amount of traffic.
Note that we still send a few periodic zero packets to avoid scenarios
where the controller fails to process the zero amplitude packet. Without
sending a few to be sure, the rumble could get stuck on until the
controller times out.
Signed-off-by: Daniel J. Ogorchock <djogorchock@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
|
|
It has been found that sending subcommands and rumble data packets at
too great a rate can result in controller disconnects. This patch limits
the rate of subcommands/rumble to once every 25 milliseconds.
Similar to sending subcommands, it is more reliable to send the rumble
data packets immediately after we've received an input report from the
controller. This results in far fewer bluetooth disconnects for the
controller.
Signed-off-by: Daniel J. Ogorchock <djogorchock@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
|
|
This patch alters the method that the rumble data is sent to the
controller. Rather than using the enable rumble subcommand for this
purpose, the driver now employs the RUMBLE_ONLY output report. This has
the advantage of not needing to receive a subcommand reply (to the major
benefit of reducing IMU latency) and also seems to make the rumble
vibrations more continuous. Perhaps most importantly it reduces
disconnects during times of heavy rumble.
Signed-off-by: Daniel J. Ogorchock <djogorchock@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
|
